Ph.D. Workshop Advanced Methods in Structural Equation Modeling Mediation and Moderation Analysis 15 th 17 th December 2015 Course content and aims: Empirical research in various disciplines (e.g., management or marketing) is replete with studies trying to establish cause and effect relationships hypothesized by the respective theories. Although finding that two variables X and Y are associated represents valuable knowledge on its own, a deeper understanding of this relationship offers both theoretical and practical benefits. More detailed insights into causal processes can be gained by answering two fundamental questions: (1) How does X influence Y, i.e. which processes and intermediate variables enable X to exert an effect on Y? (2) When does X influence Y, i.e. under what circumstances does X exert an effect on Y? Whereas how questions are approached by investigating mediation processes, analyzing moderation processes aims at answering when questions. Exploring integrated models trying to answer how and when questions simultaneously ( conditional processes ) can provide an even more finegrained picture of cause-effect relations. In this interactive course, participating Ph.D. students will learn various up-to-date statistical techniques to investigate causal hypotheses about mediation, moderation/interaction, and conditional processes in the context of regression and structural equation models with latent variables. The course will consist of a combination of lectures, exercise sessions, and a final exam (optional). During the exercise sessions following each lecture, participants will be actively involved in practical applications using Hayes SPSS macro process as well as SEM software like AMOS and R package lavaan. In addition, it will be shown how Mplus can be used to analyze interaction and quadratic effects in structural equation models. Upon completion of the course, participating Ph.D. students will understand the differences between mediating/moderating/conditional (Me/Mo/Co) processes. be familiar with state-of-the-art statistical techniques to analyze Me/Mo/Co processes. be able to appropriately interpret the analysis results for Me/Mo/Co processes. know how to best report the analysis results for Me/Mo/Co processes. be able to use various software programs (e.g., SPSS macros process, AMOS, R package lavaan, Mplus) amenable of analyzing Me/Mo/Co processes.
Language: English Place: Saarland University, Campus, building B4.1, 1st floor, room 1.03 (CIP-Pool) Course schedule: 15/12/2015 13:30 15:00 Lecture 1: Introduction to mediation, moderation and conditional processes / Mediation models with observed variables 15:15 16:45 Exercise session 1: Using SPSS macro process for mediation models with observed variables 16/12/2015 09:00 10:30 Lecture 2: Recap structural equation modeling with latent variables 10:30 10:45 Short break 10:45 12:15 Exercise session 2: Using AMOS and R package lavaan for structural equation models 12:15 13:30 Lunch break 13:30 15:00 Lecture 3: Mediation models with latent variables 15:15 16:45 Exercise session 3: Using AMOS and R package lavaan for mediation models with latent variables 17/12/2015 09:00 10:30 Lecture 4: Moderation analysis with moderated multiple regression 10:30 10:45 Short break 10:45 12:15 Exercise session 4: Using SPSS macro process for moderated multiple
regression 12:15 13:30 Lunch break 13:30 15:00 Lecture 5: Moderation analysis with latent variables / Conditional process models 15:15 16:45 Exercise session 5: Using AMOS (or R package lavaan) and Mplus for moderation models with latent variables / conditional process models Examination: At the end of the workshop, a 90 minutes exam will be offered. Participants who take the exam will receive a data set as well as a task sheet. Each task contains a graphical model presenting mediation, moderation or conditional processes for a set of variables included in the data set. In order to answer the specific questions related to the model (e.g., concerning the size and significance of a particular indirect effect in a mediation model or conditional effect in a moderation model), participants are required (1) to decide on the software appropriate for analyzing the model, (2) to make the necessary data preparations, (3) to specify the correct input file, (4) to interpret the analysis results and (5) to report and discuss the empirical findings in a written report. Participants will be graded based on the quality of the written report. Please note: Guest students only get a confirmation of participation. ECTS: The course (including the exam) is eligible for 6 credit points. Instructor: Univ.-Prof. Dr. Dirk Temme (Bergische Universität Wuppertal, Schumpeter School of Business and Economics) Contact: temme@wiwi.uni-wuppertal.de Course readings: Introduction to mediation, moderation, and conditional processes 1, pp. 3-22. Wu, A.D./Zumbo, B.D. (2008), Understanding and Using Mediators and Moderators, Social Indicators Research, 87(3), pp. 367-392. Mediation models with observed variables Hayes, A.H. (2013), Introduction to Mediation, Moderation, and Conditional Process Analysis, Chapters 4, 5, & 6, pp. 85-202.
Zhao, X., Lynch, J. G., Chen, Q. (2010), Reconsidering Baron and Kenny: Myths and Truths about Mediation Analysis, Journal of Consumer Research, 37(August), pp. 197-206. Recap structural equation modeling with latent variables Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.E. (2010), Multivariate Data Analysis A Global Perspective, 7 th ed., Upper Saddle River, NJ: Pearson, Chapter 11, pp. 629-686. Mediation models with latent variables Cheong, J.W., MacKinnon, D.P. (2013), Mediation/Indirect Effects in Structural Equation Modeling, in: Rick, R.H. (ed.), Handbook of Structural Equation Modeling, New York: Guilford, pp. 417-435. Cheung, G. W., Lau, R. S. (2008), Testing Mediation and Suppression Effects of Latent Variables: Bootstrapping With Structural Equation Models, Organizational Research Methods, 11(2), pp. 296-325. Iacobucci, D., Saldanha, N., Deng, X. (2007), A Meditation on Mediation: Evidence That Structural Equations Models Perform Better Than Regressions, Journal of Consumer Psychology, 17(2), pp. 140-154. Moderation analysis with moderated multiple regression Hargens, L.L. (2009), Product-Variable Models of Interaction Effects and Causal Mechanisms, Social Science Research, 38(1), pp. 19-28. 7, pp. 207-244. Spiller, S.A., Fitzsimons, G.J., Lynch, J.G., McClelland, G.H. (2013), Spotlights, Floodlights, and the Magic Number Zero: Simple Effects Tests in Moderated Regression, Journal of Marketing Research, 50(2), pp. 277-88. Moderation analysis with latent variables Kelava, A., Werner, C.S., Schermelleh-Engel, K., Moosbrugger, H., Zapf, D., Ma, Y., Cham, H., Aiken, L.S., West, S.G. (2011), Advanced Nonlinear Latent Variable Modeling: Distribution Analytic LMS and QML Estimators of Interaction and Quadratic Effects, Structural Equation Modeling, 18(3), pp. 465-491. Marsh, H.W., Wen, Z., Nagengast, B., Hau, K.-T. (2013), Structural Equation Models of Latent Interactions, in: Rick, R.H. (ed.), Handbook of Structural Equation Modeling, New York: Guilford, pp. 436-458.
Conditional process models 10, pp. 325-355. Attendees and application: The course is open to a maximum of 15 participants. Ph.D. students will be admitted until the maximum number of participants has been reached. Please send an email to Dr. Eva Pohl (e.pohl@eiabm.de), including your name, your thesis topic (your university, if not Saarland University) and a valid email address. Deadline for registrations: Wednesday, 1 st December 2015 Fees: The workshop is free for Ph.D. students of Faculty 1 of Saarland University. All other participants have to pay a fee of 200,- for the workshop. Furthermore, the registration as guest student (= Gasthörer) is required and a fee of 76.30 will be charged. More details at http://www.uni-saarland.de/einrichtung/zell/allgemeine-informationen-zureinschreibung/unt erlagen.html Please send us the copy of your bank receipt. For further information: Dr. Eva Pohl Tel: +49(0)681/302-2553 e.pohl@eiabm.de The European Institute for Advanced European Behavioural Management () has a tradition of more than 25 years now. The research and teaching focus is on behaviour of people, in their role as consumer, customer, employee, decision maker and markets from an economic, societal and psychological perspective. The institute offers the MBA European Management for (international) professionals working both with Europe and within Europe. The excellent choice of topics and outstanding lecturers prepare the graduates to make a remarkable contribution to the success of national and global companies.